arcanus

54 ms·

511.
by arcanus 10y ago
> biological engineering or hard AI Why not both? The future is synthesis, imo.
512.
by arcanus 10y ago
I'm sitting in a coffee shop where no one is interacting with each other, and are all staring at glowing screens of varying sizes. I'm ssh'ed into both my workstation and one of the ten most powerful supercomputers ever cons
513.
by arcanus 10y ago
> that would help consultants with their magical game-theory software make money off gullible companies / governments Yep. Game theory is also a great complex for generating academic papers, think tank analyses, etc. In reality the
514.
by arcanus 10y ago
^ This. If you're good at something, never do it for free.
515.
by arcanus 10y ago
> I enjoyed the post, thanks for writing. Same. My quibble was just part of the discussion, not intended to imply it was a poor article.
516.
by arcanus 10y ago
Interesting post! A quibble: > If you’re a Bayesian, or use a model with confidence intervals baked in, you may be in pretty good shape. But let’s face it, Bayesian techniques assume your prior is correct, and that new points are drawn f
517.
by arcanus 10y ago
I also like metal, but it is not yet performant for high performance computing, which is more my wheelhouse. I'm also skeptical it will be popular if it does not get picked up by the GPGPU folks, but time will tell.
518.
by arcanus 10y ago
I think that the closed nature of CUDA will be its undoing. I think that a standard, like C++ amp or openMP-4.5 will be the the ultimate winner. I liked openCL but it seems to be dying.
519.
by arcanus 10y ago
Almost certainly a questionable assertion given how many of the big players (FB, goog) are already locked out!
520.
by arcanus 10y ago
I'm not sure that is a problem...historically the department of Energy (DOE) has lead Supercomputing efforts in the USA and they have focused on other domains of interest (particularly nuclear weapons research). This field is actually
521.
by arcanus 10y ago
Another sign the world is moving towards a cyberpunk future, with extremely unevenly distributed outcomes. Work in tech as a full time employee at a megacorp or well funded start up and in many ways your value has never been higher. On the
522.
by arcanus 10y ago
I always wonder if in an increasingly automated world, these sort of mishaps will become more common, and take longer to elevate to a human level where it can be fixed.
523.
by arcanus 10y ago
Get the doctorate. Don't feed the trend if you don't want it to grow.
524.
by arcanus 10y ago
> The author is one of the heavyweights in numerical linear algebra (which a lot of hpc boils down to in the end), and certainly knows his stuff Came here to say this. I know Victor, and he is top notch.
525.
by arcanus 10y ago
And at what point does this become an overdiagnosis, versus an increase in prevalence? I'm by no means suggesting that autism can be an extremely serious disorder. The question is if these numbers represent an increase in serious cases
526.
by arcanus 10y ago
Is it built on closed source (nvidia cuda) so that you are locked into a particular vendor? Or something like openACC or openCL so you have a choice of accelerator (e.g. AMD or nvidia)?
527.
China to Develop Exascale Prototype in 2017(insidehpc.com)
1 points by arcanus 10y ago|0 comments
528.
by arcanus 10y ago
Two big forces, which are somewhat distinct: 1) exascale, e.g. large scale scientific compute for government and academic interests 2) hyperscale, e.g. large scale compute for industry, particularly machine learning and big data analytics W
529.
by arcanus 10y ago
Game theory is a fun way to tell stories... but it is seldom predictive. Conversely, machine learning seldom provides a story (at least in a human readable form) but it is predictive by nature.
530.
The jobs that really smart people avoid(washingtonpost.com)
58 points by arcanus 10y ago|45 comments
531.
Equipping people to stay ahead of technological change(economist.com)
4 points by arcanus 10y ago|0 comments
532.
by arcanus 10y ago
This is pretty common at all the top Engineering departments, nowadays. My doctorate was at Texas, and it certainly catered primarily to what brings in the $$: e.g. research.
533.
by arcanus 10y ago
> There is probably a PhD thesis in developing a calculus that can transform an algorithm P from one to the other architecture. Far more than a single thesis. That is a grand challenge problem.
534.
One in Three Tech Workers Is Underpaid(wsj.com)
3 points by arcanus 10y ago|1 comments
535.
by arcanus 10y ago
> it would be a win/win to be able to use TF and other similar frameworks on AMD hardware. Agree. Competition is good for the consumer. A cuDNN port that runs on AMD hardware should be available this year: http://www.anan
536.
by arcanus 10y ago
Because it was a golden age for hedge fund managers, not their clients!
537.
Carmakers Are Learning Japan’s Tech Talent Doesn’t Come Cheap(bloomberg.com)
112 points by arcanus 10y ago|158 comments
538.
by arcanus 10y ago
I wouldn't count on intel to do anything on this front until AMD proves they are competitive with zen
539.
by arcanus 10y ago
Missing the biggest triumph of 2016 by far: the victory by DeepMind's machine learning algorith at Go over the greatest living human player. At this time last year it was highly questionable if humans would ever lose to algorithms at g
540.
by arcanus 10y ago
> He used to find famous people in parties and events and never recognise them or know why they were famous. That sounds like a feature, not a bug to me.

More ›